Background: Autologous stem cell transplant is widely performed for the management of select patients with AL amyloidosis in the United States. There are no phase III trials that document improved overall survival. This phase III trial allowed patients to select between melphalan and dexamethasone or autologous stem cell transplantation. Methods: Eighty-nine patients with biopsy-proven AL amyloidosis elected to receive melphalan and dexamethasone (n = 34) or stem cell transplantation (n = 54). Patients were selected to exclude those for whom cardiac involvement would not have permitted safe autologous stem cell transplantation. Results: Allowing patients to select their preferred treatment resulted in an imbalance of patients between the two arms. Patients that selected melphalan and dexamethasone were younger, were more likely to be ECOG performance status less than 2, and had a higher amyloid stage defined by increasing degrees of cardiac amyloidosis. Patients receiving melphalan and dexamethasone had a three-year progression-free survival of 29% and overall survival of 58.8%. Patients receiving stem cell transplant had a three-year progression-free survival of 51.7% and overall survival of 83.6%. An attempt to match patients between the two arms, in terms of risk, produced 24 matched triplets with no difference in hematologic response. Conclusions: This trial, which did not meet its accrual goals, failed to demonstrate a survival advantage for matched patients undergoing autologous stem cell transplantation. Clinical trial information: NCT00477971